Coal mine working face roof disaster early warning method, system, device and storage medium

Abstract

The present application relates to the technical field of coal mine roof disaster early warning, and more particularly to a coal mine working face roof disaster early warning method, system, device and storage medium. The present application acquires hydraulic support working condition data and pre-processes; a construction risk identification matrix is constructed based on multiple risk factors of geological risk, technical equipment risk, management risk and mining risk, and a construction risk index is calculated; the safety valve opening ratio, the cycle resistance increasing rate, the passive down-drawing amount of the live column and the dynamic load coefficient are determined based on the pre-processed working condition data, and the influence coefficients of the mine pressure appearance degree corresponding to each are determined; the construction risk index and each mine pressure appearance degree influence coefficient are nonlinearly fused, each influence coefficient is weighted according to the corresponding index highlight influence coefficient and participates in the fusion, and a roof disaster risk index is obtained; early warning is performed according to the roof disaster risk index. The present application overcomes the limitations of traditional single factor evaluation, and realizes dynamic and accurate early warning of roof disaster risk.

Description

Technical Field

[0001] This invention relates to the field of early warning technology for roof disasters in coal mines, and in particular to a method, system, equipment, and storage medium for early warning of roof disasters in coal mine working faces. Background Technology

[0002] Coal mine roof disasters are a significant issue hindering safe and efficient mining, and are widely applied in conditions involving thick, hard roofs and high-intensity mining. Related technologies have constructed a roof disaster prevention and control system through the collaborative operation of overburden structure models, mine pressure monitoring, and experience-based decision-making. Specifically, this system covers the entire process from geological condition analysis to support condition assessment, including key aspects such as "transfer rock beams," "masonry beams," and "critical layer theory," and relies on statistical experience for decision-making and management.

[0003] However, existing roof fall disaster early warning methods directly use single-factor indicators without constructing a comprehensive assessment model that integrates multiple factors. This may lead to delayed disaster perception or low early warning accuracy, thus affecting safe and efficient mine production. Specifically, existing methods are difficult to obtain parameters, computationally complex, lack practical guidance, and fail to utilize hydraulic support operating data effectively, thus failing to achieve real-time and accurate early warning. Summary of the Invention

[0004] Therefore, the technical problem to be solved by the present invention is to overcome the problems of low early warning accuracy, poor real-time performance, and difficulty in comprehensively reflecting the comprehensive risk of roof disasters in the existing roof disaster early warning methods, which rely on single-factor indicators and lack a dynamic evaluation model that integrates multiple factors.

[0005] To solve the above-mentioned technical problems, the present invention provides a method for early warning of roof disasters in coal mine working faces, comprising:

[0006] Acquire the operating condition data of the hydraulic support and preprocess the operating condition data; Based on multiple risk factors including geological risk, technical equipment risk, management risk, and mining risk, a construction risk identification matrix is ​​constructed, and a construction risk index is calculated based on the construction risk identification matrix. Based on the preprocessed working condition data, the influence coefficient of the degree of mine pressure manifestation is determined. The influence coefficient of the degree of mine pressure manifestation includes the influence coefficients corresponding to the safety valve opening ratio, circulation resistance increase rate, passive retraction of the piston rod and dynamic load coefficient, respectively. The construction risk index and the influence coefficient of the degree of mine pressure manifestation are nonlinearly fused to obtain the roof disaster risk index. In the nonlinear fusion, each influence coefficient of the degree of mine pressure manifestation is weighted according to the corresponding prominent influence coefficient and then participates in the fusion. Roof disaster early warning is carried out based on the roof disaster risk index.

[0007] Preferably, the preprocessing of the operating condition data includes: Data cleaning was performed on the collected hydraulic support pressure, mining height, and piston reduction data to remove outliers and missing values, resulting in cleaned operating data. The cleaned operating data is then subjected to timestamp alignment and format standardization to generate a standardized time-series operating data dataset.

[0008] Preferably, the data cleaning of the collected hydraulic support pressure, mining height, and piston retraction data includes: The validity of pressure data, mining height data, and piston reduction data is verified based on a preset threshold range. Data exceeding the threshold range is marked as outliers and removed. Linear interpolation was used to fill the data gaps caused by outlier removal, resulting in a continuous and complete sequence of operating condition data.

[0009] Preferably, the construction risk identification matrix based on multiple risk factors, including geological risk, technical equipment risk, management risk, and mining risk, includes: Based on the actual geological conditions of the coal mine working face, geological risk factors, technical equipment risk factors, management risk factors, and mining risk factors are selected from the pre-set roof risk factor identification library. The selected risk factors are then modified by removing unimportant factors and adding important factors that affect construction safety, resulting in a modified set of risk factors. According to the advance measurement, each risk factor in the modified risk factor set is scored with a risk index, and the risk index score corresponding to each advance measurement is entered into the construction risk identification matrix.

[0010] Preferably, the construction risk index is obtained by weighted summation of the average risk index scores of various risk factors, wherein the sum of the weights of various risk factors is 1.

[0011] Preferably, determining the influence coefficient of the degree of mine pressure manifestation includes: calculating the safety valve opening ratio, circulation resistance increase rate, passive reduction of the piston rod and dynamic load coefficient based on the pre-processed working condition data, and determining the corresponding influence coefficient of the degree of mine pressure manifestation according to the activation conditions of each factor.

[0012] Preferably, the method for determining the influence coefficient of the mine pressure manifestation degree corresponding to the opening ratio of the safety valve is as follows: The first number of cycles in which the actual opening pressure of the safety valve reaches the set proportion of the designed opening pressure is counted. The first ratio of the first number of cycles to the total number of cycles is calculated. When the first ratio reaches the first threshold, the influence coefficient of the mine pressure manifestation degree corresponding to the opening proportion of the safety valve is determined according to the first ratio. Alternatively, count the number of second cycles in which the actual opening pressure of the safety valve reaches the set proportion of the designed opening pressure and the opening time reaches the set time, calculate the second ratio of the second cycle number to the total number of cycles, and when the second ratio reaches the second threshold, determine the mine pressure manifestation degree influence coefficient corresponding to the opening proportion of the safety valve based on the second ratio.

[0013] Preferably, the method for determining the influence coefficient of the degree of mineral pressure manifestation corresponding to the cyclic resistivity is as follows: Calculate the ratio of the resistance increment to the time increment of the hydraulic support in one working cycle. When the ratio reaches a set rate threshold, determine to activate the cycle resistance increase rate influence factor, and determine the mine pressure manifestation degree influence coefficient corresponding to the cycle resistance increase rate based on the number of influence factors that meet the conditions.

[0014] Preferably, the method for determining the influence coefficient of the degree of mineral pressure manifestation corresponding to the passive shrinkage of the live column is as follows: Monitor the single passive retraction amount of the hydraulic support column. When the single passive retraction amount of the column exceeds the set retraction amount threshold, determine to activate the passive retraction amount influence factor of the column, and determine the mine pressure manifestation degree influence coefficient corresponding to the passive retraction amount of the column based on the number of influence factors that meet the conditions.

[0015] Preferably, the method for determining the influence coefficient of the degree of mineral pressure manifestation corresponding to the dynamic load coefficient is as follows: The dynamic load factor is obtained by calculating the ratio of the end resistance of the cycle during the top plate pressure cycle to the end resistance of the cycle during the top plate stabilization cycle. When the dynamic load coefficient is in the first preset range, the influence coefficient of the degree of mineral pressure manifestation corresponding to the dynamic load coefficient is taken as the first preset value. When the dynamic load coefficient is in the second preset range, the influence coefficient of the degree of mineral pressure manifestation corresponding to the dynamic load coefficient is taken as the second preset value, wherein the risk level of the second preset range is higher than that of the first preset range.

[0016] Preferably, the nonlinear fusion of the construction risk index and the influence coefficient of the degree of mine pressure manifestation to obtain the roof disaster risk index includes: The influence coefficients of each mine pressure manifestation degree are exponentially calculated using the corresponding prominent influence coefficients of the indicators as exponents. The results of each exponentiation are summed, and the maximum value of the summation result is used as the exponent for square root calculation. The square root result is multiplied by the construction risk index to obtain the roof disaster risk index.

[0017] Preferably, the roof disaster early warning based on the roof disaster risk index includes: The roof disaster risk index is compared with a preset warning threshold. When the roof disaster risk index exceeds the warning threshold, a roof disaster warning signal is generated, and the main risk factors affecting the stability of the working face roof are identified, and targeted risk control measures are output.

[0018] The present invention also provides a coal mine working face roof disaster early warning system, comprising: The data acquisition and preprocessing module is used to acquire the working condition data of the hydraulic support and preprocess the working condition data. The construction risk identification module is used to construct a construction risk identification matrix based on multiple risk factors such as geological risk, technical equipment risk, management risk, and mining risk, and to calculate the construction risk index based on the construction risk identification matrix. The mine pressure manifestation analysis module is used to determine the influence coefficient of the mine pressure manifestation degree based on the preprocessed working condition data. The influence coefficient of the mine pressure manifestation degree includes the influence coefficients corresponding to the safety valve opening ratio, circulation resistance increase rate, passive reduction of the piston rod and dynamic load coefficient, respectively. The comprehensive evaluation and early warning module is used to nonlinearly fuse the construction risk index and the influence coefficient of the degree of mine pressure manifestation to obtain the roof disaster risk index. In the nonlinear fusion, each influence coefficient of the degree of mine pressure manifestation is weighted according to the corresponding prominent influence coefficient and then participates in the fusion. The roof disaster early warning is carried out based on the roof disaster risk index.

[0019] Preferably, the comprehensive evaluation and early warning module is further configured to: compare the roof disaster risk index with a preset early warning threshold; when the roof disaster risk index exceeds the early warning threshold, generate a roof disaster early warning signal, identify the main risk factors affecting the stability of the working face roof, and output targeted risk control measures.

[0020] The present invention also provides a coal mine working face roof disaster early warning device, comprising: Memory, used to store computer programs; A processor is used to implement the steps of the above-described method for early warning of roof disasters in coal mine working faces when executing the computer program.

[0021] The present invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the above-described method for early warning of roof disasters in coal mine working faces.

[0022] The technical solution of the present invention has the following advantages compared with the prior art: The coal mine working face roof disaster early warning method described in this invention constructs a construction risk identification matrix by integrating multiple risk factors such as geological risk, technical equipment risk, management risk, and mining risk. Based on real-time hydraulic support operating data, it extracts multi-dimensional mine pressure manifestation influence coefficients, including safety valve opening ratio, circulation resistance increase rate, passive retraction of the piston rod, and dynamic load coefficient. Furthermore, it nonlinearly weights and fuses the construction risk index with each influence coefficient according to the corresponding prominent influence coefficient, thereby constructing a comprehensive assessment model that fully reflects the dynamic risk of the roof. This scheme overcomes the shortcomings of traditional single-factor or linear models in depicting complex risk coupling relationships, significantly enhances the identification of key risk factors, improves the accuracy and real-time nature of early warning, and realizes dynamic quantitative assessment and immediate early warning of roof disaster risks, providing scientific and reliable decision support for coal mine safety production. Attached Figure Description

[0023] To make the content of this invention easier to understand, the invention will be further described in detail below with reference to specific embodiments and accompanying drawings, wherein: Figure 1 This is a flowchart illustrating the implementation of a method for early warning of roof collapse disasters in coal mine working faces provided by the present invention. Figure 2 This is a flowchart illustrating the implementation of a method for early warning of roof collapse in coal mine working faces, provided in one embodiment of the present invention. Figure 3 This is a schematic diagram of the process for identifying risk factors on the roof. Figure 4 This is a schematic diagram showing the roof disaster risk index and the early warning results of the support frame; Figure 5 This is a structural block diagram of a coal mine working face roof disaster early warning system provided in an embodiment of the present invention. Detailed Implementation

[0024] The core of this invention is to provide a method, system, equipment, and computer storage medium for early warning of roof disasters in coal mine working faces. It effectively integrates multiple risk factors such as geology, technology, management, and mining operations with indicators of the degree of mine pressure manifestation, and achieves dynamic and accurate early warning of roof disaster risks through a nonlinear comprehensive evaluation model.

[0025] To enable those skilled in the art to better understand the present invention, the invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. Obviously, the described embodiments are merely some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0026] Please refer to Figure 1 and Figure 2, Figure 1 This is a flowchart illustrating the implementation of a method for early warning of roof collapse in coal mine working faces provided by the present invention. Figure 2 The flowchart illustrates an embodiment of a method for early warning of roof collapse in coal mines according to the present invention; the specific operation steps are as follows: S101: Obtain the operating condition data of the hydraulic support and preprocess the operating condition data; Real-time acquisition and preprocessing of hydraulic support operating data is a fundamental step in building a roof disaster early warning system. This step aims to obtain raw information that accurately reflects the working status of the support and the dynamic response of the roof, providing reliable data support for subsequent risk identification and assessment. The acquired operating data covers various parameters that characterize the support's load-bearing characteristics, motion state, and interaction with the surrounding rock, including but not limited to the pressure values ​​of each support column, the change in support height, and the displacement of the movable column. By preprocessing this raw data, outliers introduced by sensor noise, transmission interference, or environmental fluctuations can be effectively eliminated, data formats and dimensions can be standardized, and invalid data that deviates significantly from physical laws can be removed, thereby ensuring the consistency, continuity, and accuracy of the data used in subsequent analysis. Preprocessing methods may include, but are not limited to, data filtering, outlier removal, missing value imputation, time series alignment, and normalization, and the specific method can be flexibly selected according to data characteristics and engineering requirements. As one implementation method, this step can be specifically implemented as follows: real-time acquisition of key parameters such as support column pressure, support mining height, and piston retraction amount through the electro-hydraulic control system of the hydraulic support, and preprocessing operations such as filtering and denoising, outlier removal, and timestamp alignment of the acquired raw data to form a standardized working condition data sequence.

[0027] This step, through real-time data acquisition and preprocessing, provides a high-quality and timely basic data source for roof disaster early warning, effectively reducing noise and abnormal interference in the raw data, improving the input reliability of subsequent risk analysis models, and thus laying a solid data foundation for the accurate operation of the entire early warning system.

[0028] S102: Based on multiple risk factors including geological risk, technical equipment risk, management risk, and mining risk, a construction risk identification matrix is ​​constructed, and a construction risk index is calculated based on the construction risk identification matrix; Based on multiple risk factors, including geological risk, technical and equipment risk, management risk, and mining-related risk, a construction risk identification matrix is ​​constructed, and a construction risk index is calculated based on this matrix. Specifically, this method first screens and identifies multiple risk factors affecting roof stability based on the actual geological conditions of the coal mine face, mining technology, on-site management level, and mining-related impact characteristics. These risk factors are divided into primary categories such as geological risk, technical and equipment risk, management risk, and mining-related risk, with each primary category further containing several specific sub-risk factors. Subsequently, for each advance footage, each risk factor is quantitatively scored to form a construction risk identification matrix reflecting the status of each risk factor at that advance footage. This matrix is ​​essentially a multi-dimensional dataset, with rows corresponding to different advance footage, columns corresponding to different risk factors, and elements in the matrix representing the risk index of that risk factor at the corresponding advance footage. After constructing the construction risk identification matrix, a preset fusion algorithm is used to comprehensively calculate the data in the matrix, thereby calculating the construction risk index corresponding to the current advance footage. The construction risk index is a comprehensive quantitative indicator used to characterize the potential risk level of roof slump caused by the combined effects of multiple factors, including geology, technology, management, and mining operations, at the current drilling progress. As one implementation method, the construction risk index can be calculated using the following formula: ,in, To advance the progress, Number of primary categories of risk factors For the first The impact coefficient of risk factors, and For the advance Next The risk index of each sub-factor in the class of risk factors, For the first The number of sub-factors in a class of risk factors.

[0029] This step, by constructing a construction risk identification matrix containing multiple risk factors and calculating a comprehensive construction risk index, achieves a comprehensive quantitative assessment of roof disaster risks, overcomes the limitations of traditional single-factor assessments, and provides a scientific and objective input basis for subsequent comprehensive early warning, thereby significantly improving the comprehensiveness and accuracy of the early warning model.

[0030] S103: Based on the preprocessed working condition data, determine the influence coefficient of the degree of mine pressure manifestation. The influence coefficient of the degree of mine pressure manifestation includes the influence coefficients corresponding to the safety valve opening ratio, circulation resistance increase rate, passive reduction of the piston rod, and dynamic load coefficient, respectively. Based on preprocessed operating data, a factor influencing the degree of mine pressure manifestation is calculated. This factor quantifies the dynamic changes in roof pressure intensity and support working status, serving as a key input parameter for constructing a comprehensive roof disaster evaluation model. Specifically, by statistically analyzing and dynamically monitoring real-time pressure, displacement, and time-series data collected from the hydraulic supports, mine pressure manifestation indicators across multiple dimensions are obtained, including safety valve opening ratio, circulation resistance increase rate, passive retraction of the piston rod, and dynamic load coefficient. The safety valve opening ratio reflects the frequency and extent of support overload operation. The proportion of safety valve opening cycles to total cycles and the proportion of cycles with continuous opening exceeding a set threshold are used to assess whether the roof pressure exceeds the support's rated bearing capacity. The circulation resistance increase rate characterizes the rate of resistance increase within a working cycle, used to determine the severity of roof pressure. The passive retraction of the piston rod reflects the displacement of the support's piston rod due to roof pressure; when a single retraction exceeds a preset limit, it serves as an early warning of roof pressure. The dynamic load coefficient quantifies the intensity of roof dynamic pressure impact by calculating the ratio of resistance at the end of the pressure-increasing period to resistance at the end of the stable period. As one implementation method, the values ​​of each influencing factor can be calculated based on the statistical results when the safety valve opening ratio meets preset conditions (such as the opening cycle ratio reaching the first threshold or the cycle ratio of the opening time exceeding the second threshold reaching the third threshold), the statistical results when the cycle resistance rate exceeds a certain percentage per minute of the rated working resistance, the statistical results when the single passive retraction of the piston exceeds a set value (such as 500mm), and the statistical results when the dynamic load coefficient is in different ranges (such as between the warning threshold and the danger threshold or exceeding the danger threshold), thereby providing a basis for subsequent fusion evaluation.

[0031] This step, through the collaborative calculation of multi-dimensional mine pressure manifestation indicators, can comprehensively and dynamically capture key characteristic information during the roof pressure process, overcome the one-sidedness of single indicator evaluation, significantly improve the accuracy and real-time performance of roof disaster risk identification, and provide reliable data support for the subsequent construction of a nonlinear comprehensive evaluation model.

[0032] S104: The construction risk index and the influence coefficient of the degree of mine pressure manifestation are nonlinearly fused to obtain the roof disaster risk index. In the nonlinear fusion, each influence coefficient of the degree of mine pressure manifestation is weighted according to the corresponding prominent influence coefficient and then participates in the fusion. Roof disaster early warning is carried out based on the roof disaster risk index.

[0033] After obtaining the construction risk index and the influencing factors of mine pressure manifestation, the two are fused to construct a nonlinear comprehensive evaluation model that can comprehensively reflect the dynamic risk of the working face roof. This fusion process is not a simple linear superposition, but rather, based on the different mechanisms and contributions of each influencing factor in the roof disaster evolution process, a preset fusion operator organically integrates the construction risk index with multiple mine pressure manifestation influencing factors, thereby generating a comprehensive index that can characterize the overall roof risk level under the current progress conditions. The established model has nonlinear characteristics, effectively depicting the interaction and coupling effects between various risk factors, avoiding the oversimplification of complex roof disaster processes by single indicators or linear models. Based on the output of this comprehensive evaluation model, corresponding early warning thresholds or risk level classification standards are set. When the comprehensive risk index output by the model reaches or exceeds a certain preset threshold, a roof disaster early warning signal of the corresponding level is triggered, thus providing a decision-making basis for taking targeted risk control measures on site. As one implementation method, this fusion process can adopt the following form of nonlinear comprehensive evaluation model: ,in For the advance The corresponding construction risk index, For the advance The corresponding influence coefficients of the degree of manifestation of pressure in each mine. To highlight the influence coefficient vector of the indicator, and This model can highlight the impact of key risk factors and achieve accurate early warning of roof disasters.

[0034] This technical step overcomes the shortcomings of traditional single indicators or linear models that cannot fully reflect the complex risk coupling relationship of the roof by nonlinearly integrating the construction risk index with the influencing factor of the degree of mine pressure manifestation. It significantly improves the accuracy and reliability of the early warning model, while highlighting the dominant role of key risk factors and providing more targeted early warning information for the site.

[0035] Based on the above embodiments, in some embodiments, the preprocessing of the operating condition data includes: Data cleaning was performed on the collected hydraulic support pressure, mining height, and piston reduction data to remove outliers and missing values, resulting in cleaned operating data. The cleaned operating data is then subjected to timestamp alignment and format standardization to generate a standardized time-series operating data dataset.

[0036] Furthermore, the data cleaning of the collected hydraulic support pressure, mining height, and piston retraction data includes: The validity of pressure data, mining height data, and piston reduction data is verified based on a preset threshold range. Data exceeding the threshold range is marked as outliers and removed. Linear interpolation was used to fill the data gaps caused by outlier removal, resulting in a continuous and complete sequence of operating condition data.

[0037] In this embodiment, the specific implementation of the preprocessing of the working condition data in step S101 is as follows. First, the data acquisition module acquires the raw working condition data output by the electro-hydraulic control system of the hydraulic support in real time. This raw working condition data includes at least the pressure data, mining height data, and piston retraction data of the hydraulic support. Subsequently, this raw data is cleaned. Specifically, the validity of the pressure data, mining height data, and piston retraction data is verified based on a preset threshold range. This threshold range is set according to the rated working parameters of the hydraulic support and actual working condition experience. For example, the threshold range for pressure data can be set to 0 to 1.2 times the rated working pressure, the threshold range for mining height data can be set to the minimum height to the maximum height of the support, and the threshold range for piston retraction data can be set to 0 to the maximum allowable retraction. Data exceeding the threshold range is marked as outliers and removed, thereby obtaining a pre-cleaned data sequence. For data gaps caused by removing outliers, linear interpolation is used to fill them. That is, the estimated value at the gap is calculated by using the effective data points adjacent to the gap position through a linear function, thereby obtaining a continuous and complete working condition data sequence. This continuous and complete sequence of working condition data constitutes the cleaned working condition data. Then, the cleaned working condition data undergoes timestamp alignment and format standardization. Specifically, data from different sensors or different sampling times are aligned according to a unified time base. For example, using the working cycle of the hydraulic support as the time unit, the pressure, mining height, and piston reduction data within each cycle are mapped to the same time axis. Simultaneously, the data storage format is standardized to a standardized numerical format and unit, such as pressure standardized to megapascals (MPa), mining height standardized to millimeters (mm), and piston reduction standardized to millimeters (mm), thereby generating a standardized time-series working condition dataset. This standardized time-series working condition dataset serves as input for subsequent steps S102 and S103, used to construct the construction risk identification matrix and calculate the influencing factor of the degree of mine pressure manifestation.

[0038] The above preprocessing steps effectively remove abnormal interference from the original operating data, fill in data gaps, and unify data format and time benchmark, thereby providing a high-quality and consistent data foundation for subsequent risk assessment models and significantly improving the accuracy and reliability of early warning models.

[0039] Based on the above embodiments, in some embodiments, the construction of a construction risk identification matrix based on multiple risk factors including geological risk, technical equipment risk, management risk, and mining risk includes: Based on the actual geological conditions of the coal mine working face, geological risk factors, technical equipment risk factors, management risk factors, and mining risk factors are selected from the pre-set roof risk factor identification library. The selected risk factors are then modified by removing unimportant factors and adding important factors that affect construction safety, resulting in a modified set of risk factors. According to the advance measurement, each risk factor in the modified risk factor set is scored with a risk index, and the risk index score corresponding to each advance measurement is entered into the construction risk identification matrix.

[0040] Furthermore, the construction risk index is obtained by weighted summation of the average risk index scores of various risk factors, wherein the sum of the weights of various risk factors is 1.

[0041] like Figure 3In this embodiment, the specific process of constructing the construction risk identification matrix in step S102 is as follows. First, the system reads data from a preset roof risk factor library. This library pre-stores multiple basic risk factors covering four major categories: geological risk, technical equipment risk, management risk, and mining risk. Based on the actual geological conditions of the current coal mine working face, such as coal seam thickness, roof lithology, and fault development, the system automatically or with manual assistance selects risk factors applicable to the mine from this library. For example, for geological risks, factors such as faults, fissures, and coal seam dip angle can be selected; for technical equipment risks, factors such as support working resistance saturation rate and equipment aging degree can be selected; for management risks, factors such as equipment maintenance cycle and personnel operation standardization can be selected; and for mining risks, factors such as coal pillars left in the upper goaf and mining stress concentration can be selected. Subsequently, the selected risk factors were revised, specifically including: removing factors deemed insignificant to the stability of the mine's roof, such as removing the "crack" factor in working faces with good roof integrity; and adding important factors affecting the mine's construction safety based on historical accident data or expert experience, such as adding the "frequency of microseismic events" factor in mines prone to rockbursts, thus obtaining a revised set of risk factors. This revision process ensured the relevance and effectiveness of the risk factors. Next, the system scored the risk index for each risk factor in the revised risk factor set according to the advance footage *s*. During the scoring process, multiple experts independently score the risk level of each factor at a specific advance length based on on-site observation data or historical statistical data. The score range is [1,2], where 1 represents no risk, (1,1.5] represents medium risk, and (1.5,2] represents high risk. The average score of all expert scores is then taken as the risk index score for that factor. The system inputs the score data of all risk factors corresponding to each advance length s into the construction risk identification matrix according to the advance length order and factor category. The rows of this matrix correspond to different advance lengths s, the columns correspond to the corrected risk factors, and the matrix elements are the risk index scores of the corresponding factors at that advance length. Finally, based on this construction risk identification matrix, the system calculates the construction risk index. Specifically, the construction risk index is calculated using the following formula: Where s represents the advance footage, and n represents the number of primary categories of roof risk factors, namely the number of the four major categories: geological risk, technical and equipment risk, management risk, and mining risk. Let be the influence coefficient of the i-th type of risk factor, and satisfy ∑ =1, Let be the risk index score of each factor in the i-th type of risk factor when the advance is s. Let represent the number of risk factors of type i. Using this formula, the system weights and sums the average scores of each type of risk factor with their corresponding weighting coefficients to obtain the comprehensive construction risk index R(s) for the current advance. This index reflects the level of roof construction risk caused by the combined effects of multiple factors such as geology, technology, management, and mining.

[0042] Through the above methods, this embodiment achieves dynamic screening, correction, and quantitative scoring of roof risk factors, and calculates a comprehensive construction risk index based on a weighted average model. This process can flexibly adjust the risk factor set according to the actual conditions of the mine, avoiding the limitations of a fixed factor library. At the same time, the method of averaging scores from multiple experts reduces subjective bias, making the construction of the construction risk identification matrix more scientific and objective. This provides accurate and reliable input data for the subsequent comprehensive roof disaster evaluation model, thereby improving the pertinence and accuracy of early warning.

[0043] Based on the above embodiments, in some embodiments, determining the influence coefficient of the degree of mine pressure manifestation includes: calculating the safety valve opening ratio, circulation resistance increase rate, passive reduction of the piston rod and dynamic load coefficient based on the pre-processed working condition data, and determining the corresponding influence coefficient of the degree of mine pressure manifestation according to the activation conditions of each factor.

[0044] Furthermore, the method for determining the influence coefficient of the mine pressure manifestation degree corresponding to the opening ratio of the safety valve is as follows: The first number of cycles in which the actual opening pressure of the safety valve reaches the set proportion of the designed opening pressure is counted. The first ratio of the first number of cycles to the total number of cycles is calculated. When the first ratio reaches the first threshold, the influence coefficient of the mine pressure manifestation degree corresponding to the opening proportion of the safety valve is determined according to the first ratio. Alternatively, count the number of second cycles in which the actual opening pressure of the safety valve reaches the set proportion of the designed opening pressure and the opening time reaches the set time, calculate the second ratio of the second cycle number to the total number of cycles, and when the second ratio reaches the second threshold, determine the mine pressure manifestation degree influence coefficient corresponding to the opening proportion of the safety valve based on the second ratio.

[0045] Furthermore, the method for determining the influence coefficient of the degree of mineral pressure manifestation corresponding to the cyclic resistivity is as follows: Calculate the ratio of the resistance increment to the time increment of the hydraulic support in one working cycle. When the ratio reaches a set rate threshold, determine to activate the cycle resistance increase rate influence factor, and determine the mine pressure manifestation degree influence coefficient corresponding to the cycle resistance increase rate based on the number of influence factors that meet the conditions.

[0046] Furthermore, the method for determining the influence coefficient of the degree of mineral pressure manifestation corresponding to the passive reduction of the live column is as follows: Monitor the single passive retraction amount of the hydraulic support column. When the single passive retraction amount of the column exceeds the set retraction amount threshold, determine to activate the passive retraction amount influence factor of the column, and determine the mine pressure manifestation degree influence coefficient corresponding to the passive retraction amount of the column based on the number of influence factors that meet the conditions.

[0047] Furthermore, the method for determining the influence coefficient of the mine pressure manifestation degree corresponding to the dynamic load coefficient is as follows: The dynamic load factor is obtained by calculating the ratio of the end resistance of the cycle during the top plate pressure cycle to the end resistance of the cycle during the top plate stabilization cycle. When the dynamic load coefficient is in the first preset range, the influence coefficient of the degree of mineral pressure manifestation corresponding to the dynamic load coefficient is taken as the first preset value. When the dynamic load coefficient is in the second preset range, the influence coefficient of the degree of mineral pressure manifestation corresponding to the dynamic load coefficient is taken as the second preset value, wherein the risk level of the second preset range is higher than that of the first preset range.

[0048] In this embodiment, the specific implementation method for calculating the influence coefficient of the degree of mineral pressure manifestation in step S103 is as follows.

[0049] First, based on the pre-processed hydraulic support operating data, the system calculates four basic indicators: safety valve opening ratio, circulation resistance increase rate, passive retraction of piston rod, and dynamic load coefficient. Then, based on the activation conditions of each factor, it determines the corresponding mine pressure manifestation degree influence coefficient.

[0050] Specifically, the method for determining the influence coefficient of the mine pressure manifestation degree corresponding to the safety valve opening ratio is as follows: Count the number of first cycles in which the actual opening pressure of the safety valve reaches the set ratio of the designed opening pressure; calculate the first ratio of the first cycle number to the total number of cycles; when the first ratio reaches a first threshold, determine the influence coefficient of the mine pressure manifestation degree corresponding to the safety valve opening ratio based on the first ratio. In one embodiment, the system counts the actual opening pressure of the hydraulic support safety valve. Greater than or equal to its design opening pressure 95% of the cycle number, denoted as The system calculates the ratio of the number of cycles to the total number of cycles M. When this ratio is greater than or equal to 0.6, the safety valve opening ratio influence factor is activated. Alternatively, the system counts the number of second cycles in which the actual opening pressure of the safety valve reaches the set proportion of the designed opening pressure and the opening time reaches the set time. It then calculates a second ratio of the second number of cycles to the total number of cycles. When this second ratio reaches a second threshold, the system determines the mine pressure manifestation degree influence coefficient corresponding to the safety valve opening ratio based on this second ratio. In one embodiment, the system further counts the actual opening pressure of the safety valve. Greater than or equal to the design opening pressure 95% and the duration of operation The number of cycles that are 10 minutes or longer is denoted as . The ratio of this factor to the total number of cycles M is calculated, and the factor is activated when the ratio is greater than or equal to 0.15.

[0051] The method for determining the influence coefficient of the mine pressure manifestation degree corresponding to the cyclic drag increase rate is as follows: calculate the ratio of the resistance increment to the time increment of the hydraulic support in one working cycle; when the ratio reaches a set rate threshold, determine to activate the cyclic drag increase rate influence factor; and determine the influence coefficient of the mine pressure manifestation degree corresponding to the cyclic drag increase rate based on the number of influence factors that meet the conditions. In one embodiment, the system calculates the resistance increment Δ of the hydraulic support in one working cycle. With time increment When the ratio of the cyclic resistance increase rate is greater than 5% / min of the rated working resistance, the cyclic resistance increase rate influence factor is activated.

[0052] The method for determining the influence coefficient of the mine pressure manifestation degree corresponding to the passive retraction amount of the hydraulic support column is as follows: Monitor the single passive retraction amount of the hydraulic support column; when the single passive retraction amount exceeds a set retraction amount threshold, determine that the passive retraction amount influence factor of the hydraulic support column is activated, and determine the influence coefficient of the mine pressure manifestation degree corresponding to the passive retraction amount of the hydraulic support column based on the number of influence factors that meet the conditions. In one embodiment, the system monitors the single passive retraction amount h of the hydraulic support column; when the retraction amount h exceeds 500mm, the passive retraction amount influence factor of the hydraulic support column is activated.

[0053] The dynamic load coefficient is determined by calculating the ratio of the final resistance of the cycle during the roof pressure inrush period to the final resistance of the cycle during the roof stabilization period, thus obtaining the dynamic load coefficient. When the dynamic load coefficient is within a first preset range, the dynamic load coefficient is set to a first preset value; when the dynamic load coefficient is within a second preset range, the dynamic load coefficient is set to a second preset value, wherein the risk level of the second preset range is higher than that of the first preset range. In one embodiment, the system calculates the final resistance of the cycle during the roof pressure inrush period. End-of-cycle resistance during the stabilization period of the top plate The ratio of the two values ​​yields the dynamic load factor. ,when When the dynamic load factor influence factor is taken as 0.5, when When the impact factor is 1, the impact factor is set to 1.

[0054] After determining the activation status of each factor, the system determines its corresponding mine pressure manifestation degree influence coefficient based on the activation conditions of each factor. Specifically, for the safety valve opening ratio and circulation resistance increase rate, the influence coefficient is the ratio of the number of eligible supports to the total number of supports; for the passive reduction of the piston rod, the influence coefficient is the ratio of the number of eligible supports to a set benchmark value, and this value does not exceed 1. In one implementation, the set benchmark value is 10; for the dynamic load coefficient, the influence coefficient is also the ratio of the number of eligible supports to the total number of supports. In this way, the system transforms each basic indicator into a unified mine pressure manifestation degree influence coefficient, providing input for the subsequent integrated evaluation model.

[0055] This specific implementation method clarifies the calculation logic and activation conditions of the factors affecting the degree of pressure manifestation in each mine, thereby achieving refined quantification of dynamic risks of the roof, improving the sensitivity and accuracy of the early warning model to key risk indicators, and thus enhancing the reliability and real-time performance of roof disaster early warning.

[0056] Based on the above embodiments, in some embodiments, the nonlinear fusion of the construction risk index and the influence coefficient of the mine pressure manifestation degree to obtain the roof disaster risk index includes: The influence coefficients of each mine pressure manifestation degree are exponentially calculated using the corresponding prominent influence coefficients of the indicators as exponents. The results of each exponentiation are summed, and the maximum value of the summation result is used as the exponent for square root calculation. The square root result is multiplied by the construction risk index to obtain the roof disaster risk index.

[0057] Furthermore, the roof disaster early warning based on the roof disaster risk index includes: The roof disaster risk index is compared with a preset warning threshold. When the roof disaster risk index exceeds the warning threshold, a roof disaster warning signal is generated, and the main risk factors affecting the stability of the working face roof are identified, and targeted risk control measures are output.

[0058] In this embodiment, the specific implementation method of establishing a nonlinear comprehensive evaluation model and issuing early warnings based on its output in step S104 is as follows.

[0059] First, obtain the construction risk index corresponding to the current advance s calculated in step S102. This index is based on a construction risk identification matrix and formula. The calculation shows that, among which The coefficients are the primary risk factor influence coefficients, and their sum is 1. Simultaneously, the influence coefficient of the mine pressure manifestation degree corresponding to the current advance s calculated in step S103 is obtained. These coefficients correspond to the quantitative values ​​of indicators such as the safety valve opening ratio, circulation resistance rate, passive retraction of the piston, and dynamic load coefficient. In one implementation, the specific calculation method is based on the activation conditions and formulas in Table 1: Table 1 Factors affecting the degree of mine stress manifestation v Calculation method

[0060] Note: In the stent coefficient, Σ represents the number of stents that meet the conditions, and M is the total number of stents.

[0061] The safety valve opening ratio is determined when the corresponding activation conditions are met (in one embodiment, to meet the following conditions). or The ratio of the number of eligible supports to the total number of supports M is taken; the cyclic drag ratio is determined when the corresponding activation conditions are met (in one embodiment, the cyclic drag ratio = Δ / Δt) is the ratio of the number of eligible supports to the total number of supports M; the retraction amount of the movable column is a ratio with a set number as the denominator when the single retraction amount exceeds the set retraction amount threshold (e.g., 500mm), for example, min(1,Σ / 10), where Σ represents the number of eligible supports; dynamic load coefficient Different preset values ​​are taken within different preset intervals, and the ratio of the number of stents meeting the conditions to the total number of stents M is calculated. For example, in one embodiment, when 1.2 < When ≤1.5, take 0.5; when When the value is greater than 1.5, take the value 1 and calculate Σ / M.

[0062] Then, the above construction risk index Influence coefficient on the degree of manifestation of pressure in each mine Input a nonlinear comprehensive evaluation model, which performs fusion calculations using the following formula to obtain the roof disaster risk index f(s):

[0063] in, To highlight the influence coefficient vector of the indicator, each component The factors are determined based on their relative importance in roof hazard assessment, such as in one implementation where each component... It was determined by an improved analytic hierarchy process, and This involves taking the maximum value in the vector as the base denominator for the exponential operation, thus highlighting the impact of key indicators during the fusion process. The model uses an exponentially weighted sum to non-linearly couple construction risk with the degree of mine pressure manifestation, amplifying the contribution of high-risk factors to the final risk index and achieving accurate risk quantification.

[0064] After obtaining the roof slab disaster risk index f(s), it is compared with a preset warning threshold. This warning threshold is set based on historical roof slab collapse cases and engineering experience; for example, in one implementation, it may be set to 1.5. When f(s) exceeds this warning threshold, the system automatically generates a roof slab disaster warning signal (e.g., Figure 4 As shown in the image, this indicates a risk of frame compression. Simultaneously, the system analyzes each sub-item... The contribution of the factors is used to identify the factors influencing the degree of mine pressure manifestation that contribute the most to f(s), and these factors are identified as the main risk factors affecting the stability of the current working face roof. For example, in one implementation, if the reduction in the volume of the live pillar corresponds to... If the reduction in the support column is significantly higher than other sub-items, then the reduction in the support column is determined to be the main risk factor. Finally, based on the identified main risk factor, the system matches and outputs targeted control suggestions from the preset risk control measure library. For example, in one implementation, control suggestions such as adjusting the working resistance of the support, strengthening roof support, or controlling the coal mining speed can be output to achieve accurate early warning and effective prevention and control of roof disasters.

[0065] This embodiment introduces an indicator to highlight the influence coefficient vector λ, achieving dynamic weighting of key risk factors in a nonlinear fusion model. This allows the early warning results to more sensitively reflect changes in dominant risks, improving the accuracy and relevance of the warnings. Simultaneously, by combining the identification of major risk factors with the output of control measures, it provides actionable decision support for on-site operations, effectively reducing the underreporting and false alarm rates of roof collapse disasters.

[0066] Please refer to Figure 5 , Figure 5 A structural block diagram of a coal mine working face roof disaster early warning system provided in an embodiment of the present invention; the specific system may include: The data acquisition and preprocessing module 100 is used to acquire the working condition data of the hydraulic support and preprocess the working condition data. The construction risk identification module 200 is used to construct a construction risk identification matrix based on multiple risk factors such as geological risk, technical equipment risk, management risk and mining risk, and to calculate the construction risk index based on the construction risk identification matrix. The mine pressure manifestation analysis module 300 is used to determine the influence coefficient of the mine pressure manifestation degree based on the preprocessed working condition data. The influence coefficient of the mine pressure manifestation degree includes the influence coefficients corresponding to the safety valve opening ratio, circulation resistance increase rate, passive reduction of the piston rod and dynamic load coefficient, respectively. The comprehensive evaluation and early warning module 400 is used to nonlinearly fuse the construction risk index and the influence coefficient of the degree of mine pressure manifestation to obtain the roof disaster risk index. In the nonlinear fusion, each influence coefficient of the degree of mine pressure manifestation is weighted according to the corresponding prominent influence coefficient and then participates in the fusion. The roof disaster early warning is carried out based on the roof disaster risk index.

[0067] Based on the above embodiments, in some embodiments, the comprehensive evaluation and early warning module is further used to: compare the roof disaster risk index with a preset early warning threshold, generate a roof disaster early warning signal when the roof disaster risk index exceeds the early warning threshold, identify the main risk factors affecting the stability of the working face roof, and output targeted risk control measures.

[0068] The coal mine working face roof disaster early warning system of this embodiment is used to implement the aforementioned coal mine working face roof disaster early warning method. Therefore, the specific implementation of the coal mine working face roof disaster early warning system can be found in the embodiment section of the coal mine working face roof disaster early warning method above. For example, the data acquisition and preprocessing module 100, the construction risk identification module 200, the mine pressure manifestation analysis module 300, and the comprehensive evaluation and early warning module 400 are respectively used to implement steps S101, S102, S103, and S104 in the above-mentioned coal mine working face roof disaster early warning method. Therefore, its specific implementation can be referred to the description of the corresponding embodiment, and will not be repeated here.

[0069] Based on the above embodiments, this embodiment will describe in detail the complete implementation of the aforementioned coal mine working face roof disaster early warning method, including system composition, data flow, specific implementation methods of each processing module, parameter configuration and boundary conditions, so as to further clarify the technical essence of this technical solution.

[0070] First, the coal mine working face roof disaster early warning system involved in this embodiment includes a data acquisition module, a data preprocessing module, a construction risk identification module, a mine pressure manifestation degree analysis module, and a comprehensive evaluation and early warning module. The data acquisition module is used to collect real-time operating data of the hydraulic support electro-hydraulic control system, specifically including hydraulic support pressure data, mining height data, and piston retraction data. The data acquisition module acquires raw data according to a preset sampling frequency (e.g., once per second or once per work cycle) and transmits the collected raw data to the data preprocessing module. The data preprocessing module performs data cleaning on the received raw operating data. Specifically, it verifies the validity of pressure data, mining height data, and piston retraction data based on preset threshold ranges, marking data exceeding the threshold range as outliers and removing them. For example, for pressure data, its reasonable range can be set to 0 to 120% of the rated working resistance; data exceeding this range is considered outliers. For mining height data, its reasonable range can be set to the minimum to maximum height of the support. For piston retraction data, its reasonable range can be set to 0 to the maximum stroke of the piston. For data gaps resulting from outlier removal, linear interpolation is used to fill them. This involves using adjacent valid data points to calculate the data value at the gap location using a linear function, thus obtaining a continuous and complete sequence of operating condition data. Furthermore, the cleaned operating condition data undergoes timestamp alignment and format standardization to generate a standardized time-series operating condition dataset for unified processing by subsequent modules.

[0071] The construction risk identification module constructs a construction risk identification matrix based on multiple risk factors, including geological risk, technical and equipment risk, management risk, and mining risk. Specifically, based on the actual geological conditions of the coal mine working face, risk factors affecting the stability of the working face roof are selected from a pre-set roof risk factor library. This library contains multiple risk factor categories; for example, geological risk factors may include faults and fissures; technical and equipment risk factors may include support working resistance saturation rate; management risk factors may include equipment maintenance status; and mining risk factors may include coal pillars left in the upper goaf. In practical applications, the selected risk factors are modified according to the specific conditions of each mine, removing unimportant factors and adding important factors affecting the mine's construction safety, resulting in a modified risk factor set. Then, according to the advance footage, each risk factor in the modified risk factor set is scored using a risk index. The risk index ranges from [1,2], where 1 represents no risk, (1,1.5] represents medium risk, and (1.5,2] represents high risk. The scoring is conducted simultaneously by multiple experts, and the average value is taken as the final risk index score. The risk index scores corresponding to each advance are entered into the construction risk identification matrix. The rows of this matrix correspond to different advance advances, and the columns correspond to various risk factors. See Table 2 for details. Table 2 Construction Risk Identification Matrix U (example)

[0072] Note: Risk factor values ​​range from [1,2], where 1 represents no risk, (1,1.5] represents medium risk, and (1.5,2] represents high risk. The average value is calculated based on scores from multiple experts.

[0073] When the advance footage is 10.0 meters, the score for each risk factor is 1; when the advance footage is 11.6 meters, the scores for geological risk factor U13 and technical equipment risk factor U21 are 1.8, the score for technical equipment risk factor U22 is 1.45, the scores for management risk factors U31, U32, and U33 are 1.2, 1.2, and 1.3 respectively, and the score for the remaining risk factors is 1. Based on the construction risk identification matrix, the construction risk index is calculated using the following formula:

[0074] in, To advance the progress, To identify the number of primary risk factors for the roof, For the first The influence coefficient of the risk factor and satisfying , For the advance Time Risk index scores for each sub-factor within a risk factor class. For the first The number of sub-factors of a risk factor category. This formula uses a weighted average to synthesize the scores of various risk factors, resulting in a construction risk index that reflects the overall risk level of the construction process.

[0075] The mine pressure manifestation analysis module calculates the influencing factors of mine pressure manifestation based on preprocessed operating data, specifically including the safety valve opening ratio, circulation resistance increase rate, passive piston retraction, and dynamic load coefficient. The safety valve opening ratio is calculated by statistically analyzing the actual opening pressure of the hydraulic support safety valve. Greater than or equal to the design opening pressure 95% of the cycle number And calculate its relationship with the total number of loops. The ratio of the safety valve opening ratio to the safety valve opening ratio is used when the ratio is greater than or equal to 0.6. The influence coefficient of this factor is the ratio itself. Alternatively, the actual opening pressure of the safety valve can be calculated. Greater than or equal to the design opening pressure 95% and the duration of operation Number of cycles greater than or equal to 10 minutes And calculate its relationship with the total number of loops. The ratio of the safety valve opening ratio to the operating coefficient is used. When this ratio is greater than or equal to 0.15, the safety valve opening ratio influence factor is activated, and its influence coefficient is also this ratio. The calculation method for the cyclic resistance increase rate is as follows: calculate the resistance increment of the hydraulic support within one working cycle. With time increment The ratio, i.e. When the cyclic resistance increase rate exceeds 5% / min of the rated working resistance, the cyclic resistance increase rate influence factor is activated. Its influence coefficient is the ratio of the number of eligible supports to the total number of supports M, i.e. The calculation method for the passive retraction of the piston is as follows: monitor the single passive retraction of the hydraulic support piston. When the volume of the live column decreases passively in a single instance When the diameter exceeds 500mm, the passive shrinkage influence factor of the piston column is activated, and its influence coefficient is: ,in This indicates the number of supports that meet the criteria, specifically the number of supports with a lower retraction of more than 500mm. When this number exceeds 10, the influence coefficient is 1; otherwise, it is the ratio of this number to 10. The dynamic load coefficient is calculated by calculating the end-cycle resistance during the roof pressure cycle. End-of-cycle resistance during the stabilization period of the top plate The ratio of the two values ​​yields the dynamic load factor. .when When, the dynamic load factor influence factor is taken as 0.5; when When, the dynamic load factor influence factor is taken as 1; when When the condition is within the safe range, this influence factor is not activated. The final influence coefficient of the dynamic load factor is the ratio of the number of eligible supports to the total number of supports M, i.e. .

[0076] The comprehensive evaluation and early warning module integrates the construction risk index output by the construction risk identification module and the mine pressure manifestation degree influence coefficient output by the mine pressure manifestation degree analysis module to establish a nonlinear comprehensive evaluation model. Specifically, it integrates the advance footage... Corresponding construction risk index With advance Corresponding influence coefficient of mine pressure manifestation The roof disaster risk index is obtained by fusion using the following formula. :

[0077] in, To highlight the influence coefficient vector of the indicators, the weight of the influence coefficients of each mine pressure manifestation degree in the fusion process is adjusted, and This model uses exponential and square root calculations to highlight the impact of key indicators, thereby improving the accuracy of early warnings. The calculated roof hazard risk index... Compared with the preset warning threshold, when the roof disaster risk index When the warning threshold is exceeded, a roof disaster warning signal is generated. Simultaneously, based on the contribution of each risk factor, the main risk factors affecting the stability of the working face roof are identified, and targeted risk control measures are output. For example, if the fault factor among geological risk factors has the highest contribution, control measures to strengthen the support in the fault area are output; if the support working resistance saturation rate factor among technical equipment risk factors has the highest contribution, control measures to adjust the support working parameters or replace high-resistance supports are output. By comparing and analyzing the evaluation results with on-site engineering and making corrections, the parameters of the warning model can be further optimized, improving the accuracy of the warning. Taking the 7306 working face of Shandong Energy Baodian Coal Mine as an example, this working face is 1891 meters long and 293 meters wide, using ZY12000 / 22 / 45D hydraulic supports, and the observation period is from February 2025 to December 2025. Using the method described in this embodiment, multiple risks of roof collapse were predicted at the working face, with an accuracy rate of 80% and a false negative rate of 0%, effectively achieving accurate early warning of roof disasters.

[0078] This embodiment achieves accurate early warning of roof disasters in coal mine working faces through multi-indicator fusion and improved hierarchical analysis model. It can monitor roof risks in real time, and provide targeted prevention and control measures by integrating multiple risk factors such as geology, technology, management and mining. It is applicable to different coal seam conditions and working face types, and provides strong technical support for safe and efficient coal mine production.

[0079] A specific embodiment of the present invention also provides a coal mine working face roof disaster early warning device, comprising: a memory for storing a computer program; and a processor for executing the computer program to implement the steps of the above-mentioned coal mine working face roof disaster early warning method.

[0080] A specific embodiment of the present invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the above-described method for early warning of roof disasters in coal mine working faces.

[0081] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0082] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A system that specifies functions in one or more boxes.

[0083] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including an instruction set implemented in a process. Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0084] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0085] Obviously, the above embodiments are merely illustrative examples for clear explanation and are not intended to limit the implementation. Those skilled in the art will recognize that other variations or modifications can be made based on the above description. It is neither necessary nor possible to exhaustively list all possible implementations here. However, obvious variations or modifications derived therefrom are still within the scope of protection of this invention.

Claims

1. A method for early warning of roof disasters in coal mine working faces, characterized in that, include: Acquire the operating condition data of the hydraulic support and preprocess the operating condition data; Based on multiple risk factors including geological risk, technical equipment risk, management risk, and mining risk, a construction risk identification matrix is ​​constructed, and a construction risk index is calculated based on the construction risk identification matrix. Based on the preprocessed working condition data, the influence coefficient of the degree of mine pressure manifestation is determined. The influence coefficient of the degree of mine pressure manifestation includes the influence coefficients corresponding to the safety valve opening ratio, circulation resistance increase rate, passive retraction of the piston rod and dynamic load coefficient, respectively. The construction risk index and the influence coefficient of the degree of mine pressure manifestation are nonlinearly fused to obtain the roof disaster risk index. In the nonlinear fusion, each influence coefficient of the degree of mine pressure manifestation is weighted according to the corresponding prominent influence coefficient and then participates in the fusion. Roof disaster early warning is carried out based on the roof disaster risk index.

2. The method according to claim 1, characterized in that, The preprocessing of the operating condition data includes: Data cleaning was performed on the collected hydraulic support pressure, mining height, and piston reduction data to remove outliers and missing values, resulting in cleaned operating data. The cleaned operating data is then subjected to timestamp alignment and format standardization to generate a standardized time-series operating data dataset.

3. The method according to claim 2, characterized in that, The data cleaning process for the collected hydraulic support pressure, mining height, and piston retraction data includes: The validity of pressure data, mining height data, and piston reduction data is verified based on a preset threshold range. Data exceeding the threshold range is marked as outliers and removed. Linear interpolation was used to fill the data gaps caused by outlier removal, resulting in a continuous and complete sequence of operating condition data.

4. The method according to claim 1, characterized in that, The construction risk identification matrix, constructed based on multiple risk factors including geological risk, technical equipment risk, management risk, and mining risk, includes: Based on the actual geological conditions of the coal mine working face, geological risk factors, technical equipment risk factors, management risk factors, and mining risk factors are selected from the pre-set roof risk factor identification library. The selected risk factors are then modified by removing unimportant factors and adding important factors that affect construction safety, resulting in a modified set of risk factors. According to the advance measurement, each risk factor in the modified risk factor set is scored with a risk index, and the risk index score corresponding to each advance measurement is entered into the construction risk identification matrix.

5. The method according to claim 4, characterized in that, The construction risk index is obtained by weighted summation of the average risk index scores of various risk factors, where the sum of the weights of various risk factors is 1.

6. The method according to claim 1, characterized in that, The determination of the influence coefficient of the degree of mine pressure manifestation includes: based on the pre-processed working condition data, calculating the safety valve opening ratio, circulation resistance increase rate, passive reduction of the piston rod and dynamic load coefficient respectively, and determining the corresponding influence coefficient of the degree of mine pressure manifestation according to the activation conditions of each factor.

7. The method according to claim 6, characterized in that, The method for determining the influence coefficient of the mine pressure manifestation degree corresponding to the opening ratio of the safety valve is as follows: The first number of cycles in which the actual opening pressure of the safety valve reaches the set proportion of the designed opening pressure is counted. The first ratio of the first number of cycles to the total number of cycles is calculated. When the first ratio reaches the first threshold, the influence coefficient of the mine pressure manifestation degree corresponding to the opening proportion of the safety valve is determined according to the first ratio. Alternatively, count the number of second cycles in which the actual opening pressure of the safety valve reaches the set proportion of the designed opening pressure and the opening time reaches the set time, calculate the second ratio of the second cycle number to the total number of cycles, and when the second ratio reaches the second threshold, determine the mine pressure manifestation degree influence coefficient corresponding to the opening proportion of the safety valve based on the second ratio.

8. The method according to claim 6, characterized in that, The method for determining the influence coefficient of the degree of mineral pressure manifestation corresponding to the cyclic drag increase rate is as follows: Calculate the ratio of the resistance increment to the time increment of the hydraulic support in one working cycle. When the ratio reaches a set rate threshold, determine to activate the cycle resistance increase rate influence factor, and determine the mine pressure manifestation degree influence coefficient corresponding to the cycle resistance increase rate based on the number of influence factors that meet the conditions.

9. The method according to claim 6, characterized in that, The method for determining the influence coefficient of the degree of mine pressure manifestation corresponding to the passive shrinkage of the live column is as follows: Monitor the single passive retraction amount of the hydraulic support column. When the single passive retraction amount of the column exceeds the set retraction amount threshold, determine to activate the passive retraction amount influence factor of the column, and determine the mine pressure manifestation degree influence coefficient corresponding to the passive retraction amount of the column based on the number of influence factors that meet the conditions.

10. The method according to claim 6, characterized in that, The method for determining the influence coefficient of the mine pressure manifestation degree corresponding to the dynamic load coefficient is as follows: The dynamic load factor is obtained by calculating the ratio of the end resistance of the cycle during the top plate pressure cycle to the end resistance of the cycle during the top plate stabilization cycle. When the dynamic load coefficient is in the first preset range, the influence coefficient of the degree of mineral pressure manifestation corresponding to the dynamic load coefficient is taken as the first preset value. When the dynamic load coefficient is in the second preset range, the influence coefficient of the degree of mineral pressure manifestation corresponding to the dynamic load coefficient is taken as the second preset value, wherein the risk level of the second preset range is higher than that of the first preset range.

11. The method according to claim 1, characterized in that, The nonlinear fusion of the construction risk index and the influence coefficient of the mine pressure manifestation degree to obtain the roof disaster risk index includes: The influence coefficients of each mine pressure manifestation degree are exponentially calculated using the corresponding prominent influence coefficients of the indicators as exponents. The results of each exponentiation are summed, and the maximum value of the summation result is used as the exponent for square root calculation. The square root result is multiplied by the construction risk index to obtain the roof disaster risk index.

12. The method according to claim 11, characterized in that, The method of providing early warning of roof disasters based on the roof disaster risk index includes: The roof disaster risk index is compared with a preset warning threshold. When the roof disaster risk index exceeds the warning threshold, a roof disaster warning signal is generated, and the main risk factors affecting the stability of the working face roof are identified, and targeted risk control measures are output.

13. A coal mine working face roof disaster early warning system, characterized in that, include: The data acquisition and preprocessing module is used to acquire the working condition data of the hydraulic support and preprocess the working condition data. The construction risk identification module is used to construct a construction risk identification matrix based on multiple risk factors such as geological risk, technical equipment risk, management risk, and mining risk, and to calculate the construction risk index based on the construction risk identification matrix. The mine pressure manifestation analysis module is used to determine the influence coefficient of the mine pressure manifestation degree based on the preprocessed working condition data. The influence coefficient of the mine pressure manifestation degree includes the influence coefficients corresponding to the safety valve opening ratio, circulation resistance increase rate, passive reduction of the piston rod and dynamic load coefficient, respectively. The comprehensive evaluation and early warning module is used to nonlinearly fuse the construction risk index and the influence coefficient of the degree of mine pressure manifestation to obtain the roof disaster risk index. In the nonlinear fusion, each influence coefficient of the degree of mine pressure manifestation is weighted according to the corresponding prominent influence coefficient and then participates in the fusion. The roof disaster early warning is carried out based on the roof disaster risk index.

14. The system according to claim 13, characterized in that, The comprehensive evaluation and early warning module is also used to: compare the roof disaster risk index with a preset early warning threshold; when the roof disaster risk index exceeds the early warning threshold, generate a roof disaster early warning signal, identify the main risk factors affecting the stability of the working face roof, and output targeted risk control measures.

15. A roof disaster early warning device for coal mine working faces, characterized in that, include: Memory, used to store computer programs; A processor is configured to execute the computer program to implement the steps of a method for early warning of roof disasters in a coal mine working face as described in any one of claims 1 to 12.

16. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of a method for early warning of roof disasters in a coal mine working face as described in any one of claims 1 to 12.

Images